GC-1 (Sobetirome)

Verified against PubChem PUG REST, CID 9862248, properties and synonym records retrieved 17 August 2026. Fields marked “not verified” are ones we could not confirm against a primary chemical database — we leave them blank rather than guess.

Identity and naming

PubChem resolves both sobetirome and GC-1 to a single record, CID 9862248. That record returns CAS 211110-63-3, molecular formula C20H24O4, molecular weight 328.4 g/mol, InChIKey QNAZTOHXCZPOSA-UHFFFAOYSA-N, and the IUPAC name 2-[4-[(4-hydroxy-3-propan-2-ylphenyl)methyl]-3,5-dimethylphenoxy]acetic acid. The synonym list carries UNII XQ31741E9Q, ChEMBL identifier CHEMBL107400, and the development codes QRX-431 and GC-1.

This is a small molecule, not a peptide. Structurally it is a diarylmethane analogue of thyroid hormone that carries no halogen atoms and lacks the amino-acid side chain present on triiodothyronine, two features that turn out to matter for its receptor preference.

One naming point is worth settling because it propagates through secondary material: the spelling Sobertime, which appears in some supplier notes, is a corruption. The database name is sobetirome, and it is both the international nonproprietary name and the USAN. The code NV1205 also refers to this compound, appearing as the intervention name in one registered trial.

The structural basis of the TRbeta preference

Thyroid hormone signalling in the heart runs largely through TRalpha, while hepatic lipid handling runs largely through TRbeta. A compound that separates the two receptor subtypes is therefore a useful probe, and this is the reason sobetirome has been used widely as a tool compound.

Bleicher and colleagues 2008 solved crystal structures of both human receptor ligand-binding domains in complex with GC-1 and with T3, and ran molecular dynamics simulations on the models. The ligand-binding pockets differ at exactly one residue: TRalpha carries serine 277 where TRbeta carries asparagine 331.

The structures show what that substitution does. In TRbeta, Asn331 forms a hydrogen bond with Arg282 that anchors that arginine in a single stable conformation interacting with the ligand carboxylate. In TRalpha, the shorter Ser277 side chain cannot provide equivalent anchoring, and Arg228 was observed in multiple conformations, some pointing away from the ligand. Simulated interaction energies put the productive carboxylate contact near -90 kcal/mol in both isoforms, but the stabilising Arg282-Asn331 bond at -19 kcal/mol against only -8 kcal/mol for Arg228-Ser277. The authors attributed the selectivity to the absence of the T3 amino group and the presence of the oxyacetic acid ester oxygen.

Lipid and whole-body metabolic findings in animals

Trost and colleagues 2000 compared GC-1 with equimolar T3 in hypothyroid mice, and ran dose-response work in hypercholesteremic rats. In the hypothyroid mice they reported that GC-1 lowered triglycerides more than T3 and lowered cholesterol comparably, while T3 but not GC-1 raised heart rate and elevated mRNA for the pacemaker channel HCN2. T3 but not GC-1 normalised heart and body weight and myosin heavy chain and Serca2 transcripts. The authors attributed the split to both receptor selectivity and preferential accumulation of the compound in liver relative to heart.

Villicev and colleagues 2007 treated adult female Wistar rats for six weeks with daily injections of T3 at 3 or 6 micrograms per 100 g body weight, or GC-1 at equimolar doses. Oxygen consumption rose 50 to 70 percent under both compounds. By dual-energy X-ray absorptiometry, control animals gained roughly 80 percent fat mass while GC-1-treated animals lost approximately 20 percent and T3-treated animals lost 70 to 90 percent. Direct carcass analysis put the fat reduction at 15 to 23 percent for GC-1 against 14 to 74 percent for T3. Individual skeletal muscle mass was reduced by T3 but only slightly by GC-1.

Central nervous system work

Baxi and colleagues 2014 reported that GC-1 promoted differentiation of oligodendrocyte progenitor cells from both rodent and human sources in vitro. Using PDGFalphaR-CreER;Rosa26-eYFP double-transgenic mice for genetic fate tracing, they reported that treatment during developmental myelination promoted oligodendrogenesis in corpus callosum, occipital cortex and optic nerve, and increased expression of the myelin proteins MBP, CNP and MAG in the same regions.

Hartley and colleagues 2020 worked in Abcd1 knockout mice, which carry the same inborn error of metabolism as X-linked adrenoleukodystrophy patients. They tested two strategies for enhancing the effect of sobetirome on very-long-chain fatty acids. A central-nervous-system-penetrant prodrug, Sob-AM2, lowered CNS very-long-chain fatty acids more potently than sobetirome itself and produced reduced peripheral exposure. Co-administering thyroid hormone with sobetirome enhanced lowering in the periphery but not in the central nervous system. The authors concluded that CNS lowering in that model is limited by a mechanistic threshold related to slow lipid turnover.

Note that the widely repeated claim that sobetirome upregulates the peroxisomal transporter Abcd2 is not stated in that paper's abstract and was not carried forward here.

Proliferation, tumour models, and the human record

The proliferative findings cut both ways and should be read together. Columbano and colleagues 2006 reported that GC-1 at 50 or 100 micrograms per 100 g body weight strongly stimulated hepatocyte proliferation in rats in the absence of tissue injury, along with massive pancreatic acinar cell proliferation, associated with a rapid rise in cyclin D1 mRNA but no change in c-jun or c-fos. They described the compound as a strong mitogen for both cell types.

Against that, Puliga and colleagues 2017 fed FVB mice 5 mg/kg GC-1 for 10 or 21 days in a Met-beta-catenin-driven hepatocellular carcinoma model and reported reduced tumour burden with decreased phospho-Met, phospho-ERK and phospho-STAT3, and no activation of beta-catenin-TCF4 signalling. Pourvali and colleagues 2022 reported inhibited tumour growth and reduced FDG uptake over eight days in a syngeneic mouse colorectal model.

Two clinical studies were registered. NCT01787578, a Phase 1 study of sobetirome in adult men with X-linked adrenoleukodystrophy, and NCT03196765, a Phase 1/2 study of NV1205 in boys with childhood cerebral adrenoleukodystrophy. Both are recorded as WITHDRAWN with enrollment of zero.

What is not known

No registered human outcome data exist. The only two studies entered in ClinicalTrials.gov, NCT01787578 and NCT03196765, were both withdrawn without enrolling anyone, so the compound has no registered efficacy, safety or pharmacokinetic readout in patients. Earlier Phase 1 work in healthy male volunteers is described in a 2016 review but no primary report of it was located and it is not in the registry, which means dose-exposure relationships, adverse-event rates and tolerability in people cannot be independently checked. The animal literature also contains an unresolved tension: the same compound was reported as a strong mitogen for hepatocytes and pancreatic acinar cells in normal rat tissue, and as growth-inhibiting in two mouse tumour models. No study reconciles those observations or establishes what determines which behaviour dominates. Long-term administration has not been characterised in any species in the retrieved literature; the longest treatment period found was six weeks in rats. Cartilage effects, which drove concern about the related compound eprotirome in dogs, have not been reported for sobetirome. Storage, solubility and handling conditions were not established from any primary source in this compilation.

Questions